Quantum Machine Learning Algorithms for AML
Today, the crime of money laundering is becoming more widespread and complex due to technological developments. This has made it imperative for supervisors to come up with effective solutions to combat money laundering on an international scale. With this requirement, organizations should improve their anti-money laundering strategies by adopting quantum computing, an advanced technology. Because quantum computers are well suited to detect money laundering activities that require complex analysis with their enormous computational power and their capacity to process data quickly and in a short time (Weinberg ve Faccia A., 2024).
Quantum computing makes use of qubits, unlike the binary bits used in conventional computers. While binary bits can only exist in one of two states, 1 or 0, qubits transcend this binary limitation and offer countless possibilities. A qubit can hold more than one possible value at a time and can cover a spectrum of values (Chen, L., 2024).
“Quantum machine learning” using quantum computing is a promising approach to accelerate the training and evaluation of machine learning models, and quantum machine learning algorithms have the potential to revolutionize data analysis by providing faster and more efficient solutions to complex problems (FasterCapital (2024). Therefore, quantum computing with machine learning algorithms with extraordinary computational power can assist financial institutions in detecting patterns that are indicative of money laundering activities. Some of these quantum machine learning algorithms and their anti-money laundering functions are listed in the table below (Weinberg ve Faccia A., 2024).
Tablo 1: The Functions of Quantum Machine Learning Algorithms in Combating Money Laundering
| QUANTUM MACHINE LEARNING ALGORITHMS | ANTI-MONEY LAUNDERING FUNCTIONS |
| (HYBRID QUANTUM-CLASSİCAL OPTIMISATION ALGORİTHMS ) | Accelerates risk modeling of large and complex portfolios by identifying suspicious flows that may indicate money laundering activities. |
| (QUANTUM K-MEANS CLUSTERİNG) | Quantum clustering algorithms automatically group money laundering activities based on transaction attributes and uncovers connections that might not be detected in traditional ways. |
| (QUANTUM PRINCIPAL COMPONENT ANALYSİS ) | This algorithm facilitates the detection of suspicious transactions by extracting meaningful features from large amounts of financial data. |
| (QUANTUM NEURAL NETWORKS) | Quantum neural networks can analyze massive financial transaction datasets much faster than traditional AI methods, quickly identifying suspicious patterns that point to money laundering. |
| (QUANTUM SUPPORT VECTOR MACHINES) | Quantum Neural Networks have the capacity to analyze large transaction datasets and can quickly detect suspicious transactions to combat money laundering. It has the capacity to analyze large transaction datasets such as Quantum Neural Networks and can quickly detect suspicious transactions to combat money laundering. |
Kaynak: (Weinberg A. I. ve Faccia A., 2024)
“Quantum machine learning algorithms”, with their ability to analyze massive transaction datasets, can quickly identify suspicious transactions that signal money laundering and uncover complex money laundering activities with great speed. Furthermore, with the speed of these algorithms, organizations can significantly increase their money laundering detection capabilities and fight money laundering more effectively (FasterCapital, 2024).
- Chen, L., Li, T., Chen, Y., Chen, X., Wozniak, M., Xiong, N., & Liang, W. (2024), “Design and analysis of quantum machine learning: a survey. Connection Science,” https://www.tandfonline.com/doi/full/10.1080/09540091.2024.2312121#d1e755, (07 Aralık 2024 ).
- FasterCapital (2024), “Quantum Machine Learning Algorithms” https://fastercapital.com/topics/quantum-machine-learning-algorithms.html (25 Aralık 2024).
- Weinberg A. I. ve Faccia A., (2024), “Quantum Algorithms: A New Frontier in Financial Crime Prevention” https://arxiv.org/html/2403.18322v1, (01 Ocak 2025 ).